ROGERS, Circuit Judge.

Plaintiff Dr. Alexander Stratienko, who created a catheter device and shared his design for that device with Defendant Cordis Corporation, brought this federal diversity suit against Cordis for misappropriation of a trade secret, wrongful benefit, and breach of contract. The district court granted summary judgment to Cordis, and Dr. Stratienko appeals. He challenges (1) the district court's reliance on self-interested declarations of Cordis employees, (2) the district court's determination that circumstantial evidence can never create a genuine issue of material fact in trade-secret cases, and (3) the district court's determination that the Tennessee tort of conversion does not extend to conversion of trade secrets.

The district court properly relied on declarations of Cordis employees notwithstanding the employees' self-interest. Also, although Tennessee law likely provides that Dr. Stratienko was entitled to an opportunity to demonstrate that Cordis used his catheter design by relying on circumstantial evidence, Dr. Stratienko failed to proffer sufficient evidence of similarity between his trade secret and Cordis' catheter to create a genuine issue of material fact concerning use by Cordis of Dr. Stratienko's secret. Summary judgment was, therefore, appropriate on his claims for misappropriation, wrongful benefit, and breach of contract. Dr. Stratienko's other contentions on appeal are without merit, and we accordingly affirm the judgment of the district court.

I.

Dr. Stratienko's device is a modified catheter used for gaining medical access to blood vessels. Cordis explained the conventional medical practice associated with catheters in the following manner, which is consistent with the description of conventional practice in Dr. Stratienko's patent:

An interventional cardiology procedure generally begins by inserting a hollow needle into the femoral artery, which is located close to the surface of the skin in the groin area. A wire called a guidewire is then inserted through the needle into the artery, and the needle is withdrawn. Another device called a sheath is then inserted over the guidewire into the artery. A sheath is a 4-9 inch hollow tube with a valve at the end which remains outside the body ("proximal end"). The valve prevents excess blood loss. The sheath remains in place throughout the procedure.

Next, a 35-40 inch hollow tube called a guiding catheter is inserted through the sheath and into the artery. The guiding catheter has a soft, flexible tip at the end that is inserted into the patient ("distal end"). The distal end positions the catheter into a specific blood vessel within the heart ("coronary artery"). The guiding catheter also remains in place throughout the procedure. Other medical devices such as angioplasty balloons and stents can be inserted through the guiding catheter and into the patient's coronary artery to treat the blockage.

Appellee's Br. at 4-5.

Dr. Stratienko submitted a patent application for a "sheath catheter." His design combines a sheath and a catheter with a preformed distal end. While the record is surprisingly unclear as to the benefits of the unified sheath-catheter, it appears that it allows a smaller puncture hole while providing necessary support during the invasive procedure. The preformed distal end is soft and flexible, which allows more precise engagement of targeted segments. His design also has holes at the end of the catheter, allowing direct delivery of x-ray contrast fluid to a segment of the targeted vessel; traditional delivery of contrast fluid must first displace blood to reach the intended segment.

On May 25, 1999, Dr. Stratienko, a Tennessee resident, sent a letter and Nondisclosure Agreement to Cordis Corporation. Cordis is a Florida corporation with its principal place of business in New Jersey, and is a subsidiary of Johnson Johnson. In August 1999, Cordis' Associate Manager of Business Development William Scheessele proposed changes to the agreement and faxed them to Dr. Stratienko. Over a period of months, the parties negotiated the terms of the agreement. Cordis deleted portions regarding remedies for breach and regarding return of the materials. The agreement specifically provided that Dr. Stratienko's information was a trade secret. The parties executed the agreement in November 1999. On December 12, 1999, Dr. Stratienko submitted his patent application and a description of his "sheath catheter" to Scheessele.

Cordis' internal written policy states that "Reviews [of submitted proposals] will be conducted by members of New Business Development, New Product Strategy, Legal, and to a lesser degree, senior members of R D. It is our goal to minimize exposing our innovative engineering talent pool to external ideas, to safeguard our interests to our own internal inventions." Scheessele provided Dr. Stratienko's submitted material to Cordis' in-house counsel Paul Coletti in mid-December 1999. Both Scheessele and Coletti declare that they discussed Dr. Stratienko's proposal, but they deny discussing the proposal with anyone else at Cordis or Johnson Johnson. Scheessele declares that he kept the proposal in his locked file cabinet at all times, but Coletti cannot remember whether he placed the documents in a locked cabinet or safe. On January 17, 2000, Cordis advised Dr. Stratienko that Cordis was not interested in his proposal. Coletti returned the related documents, at the latest, by mid-February and kept no copies.

On April 7, 2000, Cordis submitted a combined sheath and guiding catheter, which it called the Vista Brite Tip Introducer Guiding Catheter, for FDA approval. The FDA approved it on April 28, 2000. Cordis sales documents establish that Cordis manufactured and shipped "seventy-eight Vista Brite Tip IG Catheters as of May 29, 2000."

Dr. Stratienko alleges that Cordis has given varying reasons for its rejection of his proposal. According to Dr. Stratienko, Scheessele originally told him that Cordis was not interested because Dr. Stratienko's proposal was not within Cordis' product-development budget, and Scheessele later told him that the company had no interest in his invention. Scheessele stated at his deposition that the proposal did not meet Cordis' needs. Dr. Stratienko further alleges that Coletti told Dr. Stratienko's patent attorney that a U.S. Patent (No. 5,066,285, the Hillstead Patent) disclosed a combination sheath-catheter method similar to Cordis' Vista Brite Tip IG Catheter. Dr. Stratienko also stated in his affidavit that, for the first time in August 2002, Cordis informed him that another U.S. Patent (No. 5,897,497, the Fernandez Patent) disclosed the combination sheath-catheter. Cordis does not challenge that it gave Dr. Stratienko several reasons for its disinterest.

Dr. Stratienko filed suit in Hamilton County Circuit Court on November 21, 2001, and Cordis removed the case to the federal district court for the Eastern District of Tennessee in January 2002. Dr. Stratienko alleged that Cordis had misappropriated, converted, and wrongfully benefitted from his ideas; breached the Non-disclosure Agreement; and committed theft of a trade secret under Tennessee's Uniform Trade Secret Act, Tenn. Code Ann. §§ 47-25-1701 through 1709 (West 2000).

In March 2003, Cordis moved for summary judgment on all of Dr. Stratienko's claims. The primary pieces of evidence submitted were (1) deposition testimony and declarations of both Scheessele and Coletti and (2) Micheline Johnson and David Hill's expert reports submitted by Dr. Stratienko. David Hill was Dr. Stratienko's patent attorney. Both experts stated that neither the Hillstead nor the Fernandez patents "teach" the sheath-catheter design, and both determined that Dr. Stratienko's proposal shared all the "salient features" of the Vista Brite Tip IG, except for the holes at the distal end.

In October 2003, the district court granted Cordis' motion for summary judgment in full. The district court reasoned that, because this circuit requires direct evidence of use in trade-secret cases and because Dr. Stratienko offered only circumstantial evidence of use, Dr. Stratienko was unable to demonstrate that Cordis used his idea. Without evidence that Cordis used his secret, Dr. Stratienko was unable to rebut the evidence of Cordis' employees, who denied discussing his ideas with anyone else. Dr. Stratienko's claims for misappropriation, breach of contract, and wrongful benefit all failed, therefore, as a matter of law. The court also determined that there was no violation of the Uniform Trade Secret Act because Cordis' alleged theft in April and May 2000 occurred before the July 2000 effective date of the Tennessee statute. Finally, the district court held that Tennessee does not recognize a cause of action for conversion of trade secrets.

Earlier, during March 2003, Dr. Stratienko had moved for leave to amend his complaint so that he could add a patent-infringement claim involving a patent that is irrelevant to his trade-secret claims, and the district court had granted the motion in April. In February 2004, Dr. Stratienko moved for voluntary dismissal of this claim, but the court denied his request. In March 2004, Cordis moved for summary judgment, arguing that Dr. Stratienko's patent was not new because Cordis had patented a combined sheath-catheter in 1997, over a year before Dr. Stratienko filed his first patent application. This was the first time that Cordis had produced any evidence of its patented 1997 Webster Guiding Sheath. On the eve of the summary judgment hearing, Dr. Stratienko moved for voluntary dismissal with prejudice of his patent-infringement claim. The district court granted the motion in October 2004.

Dr. Stratienko appeals the district court's October 2003 grant of summary judgment in favor of Cordis. He challenges the rejection of each of his original claims, except for the claim under Tennessee's Uniform Trade Secrets Act. We affirm because (1) the district court properly considered the declarations of Cordis employees in determining that Cordis, as the moving party, satisfied its burden of demonstrating that no genuine issue of material fact existed; (2) Dr. Stratienko did not respond with sufficient evidence to create a genuine issue of material fact as to whether Cordis used his catheter design; and (3) Tennessee's cause of action for conversion does not extend to conversion of trade secrets.

II.

This court reviews the district court's grant of summary judgment de novo. See United States v. Miami Univ., 294 F.3d 797, 805 (6th Cir. 2002). Summary judgment is appropriate only when no genuine issue of material fact exists and when the moving party is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(c). This court must view the facts contained in the record and draw all inferences from the record in the light most favorable to the nonmoving party. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). This court cannot weigh the evidence or determine the truth of any matter in dispute, Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986), because this court determines only whether the case contains sufficient evidence from which a jury could reasonably find for the nonmoving party, id. at 248-49, 106 S.Ct. 2505. If the moving party fulfills its burden of demonstrating that no genuine issue of material fact exists, the nonmoving party, to receive a trial, must present some significant probative evidence creating a factual issue. Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). Because Cordis met its burden and Dr. Stratienko failed to meet his, summary judgment was proper.

III.

Cordis submitted declarations by Coletti and Scheessele, in which they deny sharing Dr. Stratienko's information with anyone else at Cordis. Although these declarations came from self-interested employees, Cordis can rely on those denials to demonstrate that no genuine issue of material fact exists because Dr. Stratienko has not impeached the declarants' credibility. Dr. Stratienko argues that the following language in Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 151, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000) (quoting 9A Charles Alan Wright Arthur R. Miller, Federal Practice and Procedure at 300 (2d ed. 1995)), precludes the district court from considering Scheessele and Coletti's declarations in deciding a motion for summary judgment and thus leaves Cordis with no evidence denying use: "[T]he court should give credence to the evidence favoring the nonmovant as well as that `evidence supporting the moving party that is uncontradicted and unimpeached, at least to the extent that that evidence comes from disinterested witnesses.'" Dr. Stratienko misinterprets the import of this language to mean that courts may never consider affidavits of interested persons when the affidavits are submitted by a moving party.

This court has already considered this issue in Almond v. ABB Indus. Sys., Inc., 56 Fed.Appx. 672, 2003 WL 173640 (6th Cir. Jan. 22, 2003) (per curiam), and held that courts can consider the testimony of a moving party's interested witnesses. The court held that the interpretation of Reeves advocated by Dr. Stratienko "leads to absurd consequences" because defendants will often be able to respond only through the testimony of their employees. Almond, 56 Fed.Appx. 672, 2003 WL 173640, at *2. To support its conclusion, the Almond court cited additional language from Federal Practice and Procedure: "The testimony of an employee of [the movant] must be taken as true when it disclosed no lack of candor, the witness was not impeached, his credibility was not questioned, and the accuracy of his testimony was not controverted by evidence. . . ." Wright Miller at 287 n. 9. Almond's holding is consistent with Chesapeake Ohio Ry. v. Martin, 283 U.S. 209, 218, 51 S.Ct. 453, 75 L.Ed. 983 (1931), in which the Supreme Court stated that courts need not deny the conclusiveness of testimony of the moving party that "is not contradicted by direct evidence, nor by any legitimate inferences from the evidence[,]" because the rule requiring that testimony be considered by the jury is not "an absolute and inflexible one." Almond and Chesapeake establish that the issue, therefore, is not whether the district court could consider the affidavits of Cordis but instead whether the affidavits were uncontradicted.

Dr. Stratienko unsuccessfully contends that the following pieces of evidence so impeach or controvert the testimony of Scheessele and Coletti that their testimony is not sufficient for summary judgment: (1) Cordis did not follow its internal policy, (2) Coletti left his office unlocked, (3) Coletti retained Dr. Stratienko's information for weeks after he declined the proposal, (4) Cordis recommended changes to the Nondisclosure Agreement, and (5) Cordis provided several reasons for declining the proposal. Because none of these facts, however, impeach Scheessele and Coletti's testimony, the district court correctly relied upon Cordis' declarations.

Cordis followed its internal policy, and thus the internal policy has no bearing on the declarants' credibility. Dr. Stratienko argues that Cordis' policy, stating that reviews "will be conducted by members of New Business Development, New Product Strategy, Legal, and to a lesser degree, senior members of R D[,]" impeaches the declarants' testimony that they were the only individuals to see his proposal. There is no inconsistency here. Cordis established the policy "to minimize exposing our innovative engineering talent pool to external ideas, to safeguard our interests to our own internal inventions." The policy does not mandate that other individuals read the proposal. Its purpose, as evidenced by "to a lesser degree," is to limit the individuals who have access to secrets. In this case, consistent with the policy, Scheessele, a member of New Business Development, and Coletti, a member of the legal department, examined the proposal. The fact that the policy may have permitted others to view the proposal does not create any evidence that they, in fact, viewed it.

Coletti's uncertainty as to whether he kept Dr. Stratienko's documents in a locked or unlocked office fails to impeach his testimony that he did not share Dr. Stratienko's information with others at Cordis. Dr. Stratienko contends that Coletti said that the documents were left in an unlocked office. But Coletti, instead, said that he was unsure whether the documents were locked away. The small possibility that the documents were stolen or viewed sub rosa by being in a potentially unlocked office does not impeach Coletti's testimony that he did not share Dr. Stratienko's information with anyone at Cordis.

Likewise, Coletti's retention of Dr. Stratienko's information for weeks after he declined the proposal does not contradict the declarations. Although Dr. Stratienko argues that Coletti's retention of the documents creates a genuine issue of material fact as to whether Cordis used the documents, the retention does not demonstrate in any way that Cordis used the information or that Coletti shared the information with anyone else. Any allowable inference that the retention of documents for a few weeks increases the probability that Cordis shared them with Research and Development is too tenuous to create a genuine issue of material fact as to use.

That Cordis had recommended changes to the Nondisclosure Agreement in no way impeaches the testimony of Scheessele and Coletti. Dr. Stratienko argues that, because Cordis' suggested changes affected the remedies and the return of documents, the declarants' credibility is impeached. Both Cordis and Dr. Stratienko negotiated the details of the agreement, and the agreement contains several provisions that protect Dr. Stratienko. Dr. Stratienko never explains how the revised terms, reached at arm's length, affect credibility or make it any more likely that Cordis used the proposal.

Dr. Stratienko's final argument — that Cordis' several reasons for declining the proposal impeach the declarants' testimony — also fails because the various reasons are not contradictory. According to Dr. Stratienko, Coletti gave only one reason: that the Hillstead patent disclosed the catheter-sheath combination. Scheessele gave three reasons: (1) that Cordis was not interested; (2) that the proposal did not meet Cordis' needs; and (3) that the proposal was not within Cordis' budget. Dr. Stratienko also alleges that Cordis later said that the Fernandez patent disclosed the catheter-sheath combination. None of these reasons is inconsistent with another. Cordis could have been uninterested because the proposal was not within its budget and did not meet its needs. Reliance on two other patents also raises no inconsistency because Cordis was not limited to only one patent from which to glean inspiration. Moreover, relying on two patents provides good reason why Cordis was uninterested. Even assuming that the recitation of several reasons creates credibility issues as to why Cordis declined the proposal, it does not make it any more likely that the proposal was given to others or used in any way. Dr. Stratienko's attempts to impeach the declarants' credibility thus are insufficient to require the conclusion that the district court could not consider the declarations submitted by Cordis in considering whether Cordis met its burden of demonstrating the absence of any genuine issue of material fact.

IV.

Because Cordis' declarations satisfied Cordis' burden, the burden of creating a genuine issue of material fact shifted to Dr. Stratienko, who failed to produce sufficient evidence of use to create a genuine issue of material fact as to whether Cordis used his secret. The district court held that Dr. Stratienko could not demonstrate that Cordis used his design through circumstantial evidence of (1) Cordis' access to Dr. Stratienko's secret and (2) the similarity between Dr. Stratienko's secret and Cordis' catheter. Even assuming that circumstantial evidence in the form of access and similarity may in some cases be sufficient evidence of use, such evidence is not sufficient here. Dr. Stratienko's evidence fails to identify which, if any, innovative features his and Cordis' designs share, and he is, therefore, unable to show sufficient relevant similarity to permit a circumstantial inference in this case. In light of the uncontested declarations of Cordis' employees and Dr. Stratienko's failure to produce sufficient evidence of shared innovative features, there is no genuine issue of material fact that Cordis used Dr. Stratienko's design.

The district court accurately described the four elements under Tennessee law for misappropriation of a trade secret: (1) the existence of a trade secret; (2) communication of the trade secret to the defendant while in a position of trust and confidence; (3) defendant's use of the communicated information; and (4) resulting detriment to the plaintiff. See Hickory Specialties, Inc. v. B L Labs., Inc., 592 S.W.2d 583, 586 (Tenn.Ct.App. 1979) (citing Smith v. Dravo Corp., 203 F.2d 369, 373 (7th Cir. 1953)). The issue in this case concerns the third element — whether Cordis used the confidential information that Dr. Stratienko communicated in his proposal. The parties have not referred to and we have not discovered any Tennessee cases considering whether parties may rely on circumstantial evidence to prove use of a trade secret. "Where the state supreme court has not spoken, our task is to discern, from all available sources, how that court would respond if confronted with the issue." Rector v. Gen. Motors Corp., 963 F.2d 144, 146 (6th Cir. 1992).

Contrary to the district court's reasoning, a strong argument could be made that courts may properly consider circumstantial evidence concerning similarity of design plus access to the design to imply use by a defendant of a trade secret. Other circuit courts have permitted such an inference in trade-secret and nondisclosure cases. See Sokol Crystal Prods., Inc. v. DSC Commc'ns Corp., 15 F.3d 1427, 1432 (7th Cir. 1994); see also Leggett Platt, Inc. v. Hickory Springs Mfg., 285 F.3d 1353, 1361 (Fed. Cir. 2002); Pioneer Hi-Bred Int'l v. Holden Found. Seeds, Inc., 35 F.3d 1226, 1239 (8th Cir. 1994); Eden Hannon Co. v. Sumitomo Trust Banking Co., 914 F.2d 556, 561-62 (4th Cir. 1990); Electro-Miniatures Corp. v. Wendon Co., 771 F.2d 23, 26 (2d Cir. 1985); SI Handling Sys. v. Heisley, 753 F.2d 1244, 1261 (3d Cir. 1985); Droeger v. Welsh Sporting Goods Corp., 541 F.2d 790, 793 (9th Cir. 1976). Sufficient circumstantial evidence of use in trade-secret cases must demonstrate that (1) the misappropriating party had access to the secret and (2) the secret and the defendant's design share similar features. See, e.g., Leggett Platt, 285 F.3d at 1361; Droeger, 541 F.2d at 793. These cases support the proposition that, once evidence of access and similarity is proffered, it is "entirely reasonable for [the jury] to infer that [defendant] used [plaintiff's] trade secret." Sokol, 15 F.3d at 1432.

Permitting an inference of use from evidence of access and similarity is sound because "[m]isappropriation and misuse can rarely be proved by convincing direct evidence." Eden Hannon Co., 914 F.2d at 561 (citing Greenberg v. Croydon Plastics Co., 378 F.Supp. 806, 814 (E.D.Pa. 1974)). Presented with "defendants' witnesses who directly deny everything," plaintiffs are often required to "construct a web of perhaps ambiguous circumstantial evidence from which the trier of fact may draw inferences which convince him that it is more probable than not that what the plaintiffs allege happened did in fact take place." Id. Thus, requiring direct evidence would foreclose most trade-secret claims from reaching the jury because corporations rarely keep direct evidence of their use ready for another party to discover. Caselaw from other circuits thus suggests that Tennessee law would most likely permit circumstantial evidence of use in trade-secret cases.

In the case at bar, we assume that Dr. Stratienko produced sufficient evidence of access: Cordis' declarations demonstrate that Cordis, the entity that developed the Vista Brite IG Catheter, had access to the information. It would be very burdensome to require Dr. Stratienko to demonstrate that a particular inventor or designer had access. See Sokol, 15 F.3d at 1432 (stating that any plaintiff "would be hard pressed to present direct proof of the flow of information inside the defendant's company").

Dr. Stratienko, however, has not presented sufficient circumstantial evidence of similarity to withstand summary judgment. His evidence does not reveal which novel features Cordis' device shares with Dr. Stratienko's device. His expert reports essentially state two facts: that (1) the Fernandez and Hillstead patents do not "teach" the features of Cordis' Vista Brite Tip IG Catheter and (2) the Stratienko device shares "all of the salient features found in the Cordis IG." Both of these conclusions are based on irrelevant comparisons. The experts' first conclusion that Cordis' catheter is dissimilar from the Fernandez and Hillstead catheters has no bearing on whether Cordis' and Stratienko's catheters are similar. The similarity inquiry is not a comparison of several products to determine which two products are the most similar. Dr. Stratienko must, instead, demonstrate similarity by comparing his secret with the features of Cordis' catheter, not the features of Cordis' catheter with the features of other catheters.

The experts' conclusions that the catheters share "all . . . salient features" except the holes at the distal end are also insufficient because their findings do not reveal which feature was the secret that Dr. Stratienko sought to protect. Misappropriation of a trade secret requires that an idea not be one that is already used by others. See Hickory Specialties, Inc., 592 S.W.2d at 586 (stating that the device must give the creator "an opportunity to obtain an advantage over competitors who do not use it"). Dr. Stratienko must demonstrate similarity between his secret idea (not his product in general) and Cordis' device. See Leggett Platt, Inc., 285 F.3d at 1361. Instead, his experts state only that the Stratienko device has a catheter-sheath combination, a preformed tip, and holes at the distal end. At no point do his experts state which, if any, of these characteristics renders his idea innovative. It is entirely possible that the holes at the distal end, which the Cordis design lacked, distinguished Dr. Stratienko's device from prior art. We cannot find in the record that Dr. Stratienko has presented evidence of which features were novel to his design, and thus he has failed to show sufficient similarity between the innovative aspect of his idea and Cordis' device to provide enough of a circumstantial inference to create a genuine issue of material fact.

It is true that the parties have already conceded that Dr. Stratienko had a trade secret. But this does not mean that sufficient similarity has been shown to permit an inference of use. The parties' implicit concession that a trade secret exists does not identify the secret. Identifying the secret that provides Dr. Stratienko with an advantage in the market is necessary for determining whether the pertinent similarity implies that Cordis used his secret. The analysis of similarity evaluates only relevant, innovative features, not all possible congruence. Without evidence of the advantage of Dr. Stratienko's device over prior art, there is not a sufficient basis for a reasonable jury to make the circumstantial inference of use. Dr. Stratienko, therefore, has not presented sufficient evidence to establish that there is any genuine issue of material fact as to use, and summary judgment was proper as to Dr. Stratienko's claims for misappropriation, wrongful benefit, and breach of contract.

V.

The district court also correctly determined that Tennessee law does not recognize a cause of action for conversion of trade secrets because no decision of the Supreme Court of Tennessee calls into doubt the state court of appeals' express holding rejecting a claim for conversion of intangible property. In B L Corp. v. Thomas Thorngren, Inc., 917 S.W.2d 674, 680 (Tenn.Ct.App. 1995), the Court of Appeals of Tennessee, after considering the decisions of other courts and determining that only a minority of courts recognizes conversion of intangible property, held that Tennessee does not recognize an action for conversion of intangible property. That court has subsequently relied upon its holding in B L. See Ralph v. Pipkin, 2005 WL 1220132, at *3 (Tenn.Ct. App. May 17, 2005); Corporate Catering, Inc. v. Corporate Catering, Etc., LLC, 2001 WL 266041, at *5 (Tenn.Ct.App. Mar. 20, 2001).

Dr. Stratienko's invocation of language in a state supreme court opinion, which he claims supports recognizing conversion of intangible property, is unavailing because he misunderstands the import of the language in that opinion. He points to the following sentence in Barger v. Webb, 216 Tenn. 275, 391 S.W.2d 664, 665 (1965) (emphasis added): "A conversion, in the sense of the law of trover, is the appropriation of the thing to the party's own use and benefit, by the exercise of dominion over it. . . ." He argues that, because "trover" applies to personal property, which includes intangible property, the Supreme Court of Tennessee recognizes conversion of intangible property. The Supreme Court of Tennessee's recognition that the modern tort of conversion grew out of the old writ for trover, however, in no way implies that the modern tort extends to intangible property. Indeed, if anything, the opposite conclusion is implied by the historical fact that trover traditionally did not extend to intangible property. See W. Page Keeton et al., Prosser Keeton on Torts 91 (5th ed. 1984). Because no opinion of the Supreme Court of Tennessee calls into doubt the state court of appeals' holding that Tennessee does not extend its cause of action for conversion to intangible property, we are unable to conclude that Tennessee law recognizes a tort for conversion of trade secrets.

VI.

For the foregoing reasons, we AFFIRM the judgment of the district court.